NanoVNA Custom Software

[joeqsmith]

I am not understanding your graph and the criteria you use to change gains.

[DiSlord]

on lower 400kHz always used gain 0 in any case for all (on enabled AGC)

Not remember why i use this (possible need enable, but at this moment i not have free time for tests).

[gf]

on lower 400kHz always used gain 0 in any case for all (on enabled AGC)
Not remember why i use this (possible need enable, but at this moment i not have free time for tests).

I'm surprised that gain 3 is used at all below 300 MHz, if it cannot be calibrated:

... but in less ¬300MHz range not possible calibrate Gain 3 due to overflow in any case.

[joeqsmith]

Shown with the 100uOhm and 1mOhm resistors using an average of 10 and gain of 3.    I think this is the first time I have been able to discern between these two parts using the Lite.

[DiSlord]

LiteVNA test firmware
I enable AGC in all range for tests
This fw also contain new traces for S21 measures (Series R/X and Shunt R/X and Q)
Small improvements for LCD draw and SD cards

[joeqsmith]

Just loaded.  It appears to be for the small display.  I'll go ahead and run it anyway. 

[DiSlord]

Archive contain 2 firmwares LiteVNA 64 for big and LiteVNA 62 for small displays

[joeqsmith]

Archive contain 2 firmwares LiteVNA 64 for big and LiteVNA 62 for small displays

I don't think I ever sorted out where you have them archived.

https://www.eevblog.com/forum/rf-microwave/nanovna-custom-software/msg3928550/#msg3928550
https://www.eevblog.com/forum/rf-microwave/nanovna-custom-software/msg3928592/#msg3928592
https://www.eevblog.com/forum/rf-microwave/nanovna-custom-software/msg3930395/#msg3930395

I never joined that groupio mail list for beta testers.  Is that the only way to gain access?

Anyway, the new firmware does appear to correct the AGC limitation.   Shown with a 1mOhm, 0.1mOhm and short.   Graph is S21, not resistance.   Not anything useful below 50kHz but certainly improves the lower region. 

[gf]

Not anything useful below 50kHz but certainly improves the lower region. 

If there is no error in my calculation¹⁾, then 1mOhm shunt has S21 of -91dB, and 0.1 mOhm -111dB.
Noise floor is still too high for that use case. Ideally, you'd rather need a noise floor of say -120dB, or even lower.
You could try to increase the averaging further to 100x or 1000x. Rather reduce the number of points, if it becomes too slow.
But even if you can further reduce the noise floor, I wonder whether the gain calibration will be accurate enough at such low levels.
The discontinuity at 400 kHz is quite large, too.

¹⁾ https://cds.cern.ch/record/1415639/files/p67.pdf, page 88, "3. shunt admittance", for Y1=Y2=1/50, and Y=1kS and Y=10kS

[joeqsmith]

Math seems in the ball park.

Some time ago when I made these two standards and transformer, I used my higher end equipment.   I am using them with the Lite only for a fringe case.   The original NanoVNA did a much better job of it.   

10x average takes a very long time to run.  One sweep maybe 6 minutes or so.  100X about an hour per sweep.   Compare with a few seconds on my vintage VNA.   

[joeqsmith]

Working on a rotor to make 3D radiation measurements. 

[joeqsmith]

5 deg resolution.   

[joeqsmith]

I was wanting to try something a little more directional on the new 3D mount and decided on putting a simple antenna together with a discarded bean can.     

https://jacobsalmela.com/2013/09/07/wi-fi-cantenna-2-4ghz-how-to-make-a-long-range-wi-fi-antenna/

The can I used measures 109mm x 73mm.  From their drawing, this can should be good for 2.1GHz, or a wavelength of 143mm.    After construction, the Lite showed very poor SWR so I added a few stubs to try and tame it. 

[joeqsmith]

CAN9
Running the can at 4GHz.

CAN10
Comparing the 4GHz radiation pattern with running the same setup at 5GHz.  It looses all directivity.

CAN11
Comparing previous data sets with the same setup at 2GHz (1.94GHz).

CAN12
Looking at the 2GHz data in 3D.

[joeqsmith]

CAN13
Moving the bean can further away and running a much higher resolution sweep.   Next time I have beans, I'll lengthen the waveguide.

***
This was a paper I had read on making these measurements.   
https://www.ets-lindgren.com/WhitePapers/APM.pdf

[joeqsmith]

Using the newer 3D plotting with the wire frame disabled.   This scan was taken at 2 degrees or 16,200 data points.   Imagine taking that by hand. 

Next step is add some sort of auto align and measurement readouts.  Maybe a basic field calculator.   

[joeqsmith]

Rather than having an auto alignment for the antenna, I just rifle through the data to find the peak and consider that head on.   Showing calcs for the half power beam width.  Both azimuth and elevation are calculated.  For the front-to-back ratio I plan to use the BASTA standard. 

Without a chamber and reference antennas, it's more just for the fun of it. 

http://www.mobi-antenna.com/uploadfiles/2017/07/201707241212321232.pdf

[metrologist]

Motors for rotor system.

Would you be willing to create a block diagram of your system setup?

[joeqsmith]

Motors for rotor system.

Would you be willing to create a block diagram of your system setup?

If you read those few links I had posted and watched the video, it should have been fairly clear but let me try and expand on it for you.

I have the PC tied to the LiteVNA using USB.   I have two M-Drive motors as my document describes fitted with a gear drive.   These are configured for multi-drop on an RS-485 bus.  One motor is used for the azimuth (horizontal) the other for elevation (vertical).   There is a separate power supply for the two motors.    The RS485 to USB adapter is connected to the PC as well.    The LiteVNA has two ports.  Port 1 connects to the AUT (antenna under test) and Port 2 to the reference antenna.    The LiteVNA sends the signal out Port 1 to the AUT.  The signal is then received by the reference antenna on Port 2.   The software tracks the signal strength while it moves the antenna. 

Currently, the software starts by pointing the AUT downwards and sweeps the azimuth from 0-360 degrees.  The elevation is then set to the next angle and the process repeats until the elevation has swept from 0-180 degrees.   This gives us our full sphere.   

I made a quick block diagram for you but it isn't very helpful as there really isn't much to it.  There are lots of papers and videos on this subject if you are interested.   Maybe the following videos will help.  I hate to guess what that whole setup costs.   

****
Block diagram shows PORT1&2 swapped to my current setup.

[metrologist]

Thanks Joe

[joeqsmith]

Shown is the Vivaldi I attempted to copy from our friend who was experimenting with RADAR.   There is also a poorly constructed dipole along with a fresh can of pork and beans.  Used the Dremel tool to cut off the bottom of the can and removed the ring from the top of the original can.  After that, it was easy enough to solder together.  I moved the mounting bracket to center the antenna. 

The dipole sort of radiates like we would expect. 

[joeqsmith]

The Vivaldi compared with the dipole. 

[joeqsmith]

Showing the new bean can antenna pointing downward compared with the Vivaldi and dipole.  Stubs had to be retuned after extending the waveguide.

[joeqsmith]

Showing the SWR after extending the waveguide and adjusting the stubs.   It seems to improve the side lobes.   
 

[FPSychotic]

I'm not able to make work the nanoVNA solver lite/v2 +4 for the liteVNA.
Anyone have it running in windows 11? I really having a bad time trying to understand the issue, no idea if it is the software that just won't run as it is deprecated, it can run but I had something wrong with the National instruments runtimes that Im unable to understand even what they are, what they do , just unable to make it work.  Other softwares works. I don't want disturb, sorry if this not proceed.